[all-commits] [llvm/llvm-project] e75e24: [AMDGPU][True16][Codegen] keep srcmod/clamp/omod f...

[Brox Chen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e75e2485f2e5e627d0bdf0306df4672f69ddd6eb
      https://github.com/llvm/llvm-project/commit/e75e2485f2e5e627d0bdf0306df4672f69ddd6eb
  Author: Brox Chen <guochen2 at amd.com>
  Date:   2025-06-19 (Thu, 19 Jun 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
    A llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-f16-gfx12-fake16.mir
    A llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-f16-gfx12-true16.mir
    M llvm/test/CodeGen/AMDGPU/move-to-valu-pseudo-scalar-trans-f16-fake16.ll
    M llvm/test/CodeGen/AMDGPU/move-to-valu-pseudo-scalar-trans-f16-true16.ll

  Log Message:
  -----------
  [AMDGPU][True16][Codegen] keep srcmod/clamp/omod from v_s_xxx_f16 when moved to VALU (#144781)

https://github.com/llvm/llvm-project/pull/141152 causes an issue in
v_s_xxx_f16 lowering in both true16/fake16 flow.

V_S_XXX_F16 are special insts which has scalar input/output but in VALU
VOP3 format. Need to keep the srcmod/clamp/omod when lower it to its
corresponding VALU inst with vector input/output.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications